Stackless Python, eventual merge?

Christian Tismer tismer at
Fri Sep 20 09:19:19 EDT 2002

Greg Ewing wrote:
> Martin v. Loewis wrote:
>> That is not true. It is more reliable to inspect the source code.
>> In general, it is very easy to do so: if there are no callbacks to
>> Python, the module is stackless-safe.
> That's a fair point, and I guess it takes care of
> the majority of extensions.
> But the cases where there is doubt are typically
> going to be big, complicated things like GUI
> libraries. Making a line-by-line inspection of
> one of those isn't exactly something you do over
> a cup of tea.
> And you don't just have to look at the code, you
> have to understand what it's doing, because the
> things which cause problems are dynamic rather
> than static features of the code.

Fine. But still I'm trying to see your true message.
Why is it bad to just switch off switching when
an unknown extension is entered?
Making that xtension stackless-safe is in fact
a project, I can't see anything bad with it.

Compare it to olde Stackless: It refused to switch
anything that wasn't the toplevel interpreter.
New Stackless can be restricted in the same way.
The only difference is that the new technology
is *able* to switch any stacks. This should only
be enabled if one knows what she's doing.

But there is no reason to damn a new technology
just because it is very powerful. Run it tamed
by default, and everything is fine and by no means
worse than before.
On the other hand, if you want to take advantage
of it with extension modules, then analyse these
modules. This is a project, and I'm greatful for
any contribution.

but-there-is-no-free-lunch -- ciao - chris
Christian Tismer             :^)   <mailto:tismer at>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Johannes-Niemeyer-Weg 9a     :    *Starship*
14109 Berlin                 :     PGP key ->
work +49 30 89 09 53 34  home +49 30 802 86 56  pager +49 173 24 18 776
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?

More information about the Python-list mailing list